Homework 9:
WordNet
Benjamin Roth, Marina Sedinkina Symbolische Programmiersprache Due: Thursday January 23, 2020, 16:00
In this exercise you will:
• measure semantic similarity of words using WordNet
• find hyponyms of the given hypernyms in the text
This homework will be graded using unit tests by running: python3 -m unittest -v hw09_wordnet/test_wordnet.py
Exercise 1: WordNet semantic similarity [8 points]
Use the predefined path-based similarity measures (accessible with the use of
synset1.path_similarity(synset2)) to score the similarity of each of the following pairs of words: car-automobile, gem-jewel, journey-voyage, boy-lad, coast-shore, asylum- madhouse, magician-wizard, midday-noon, furnace- stove, food-fruit, bird-cock, bird- crane, tool-implement, brother-monk, lad- brother, crane-implement, journey-car, monk- oracle, cemetery-woodland, food- rooster, coast-hill, forest-graveyard, shore-woodland, monk-slave, coast-forest, lad-wizard, chord-smile, glass-magician, rooster-voyage, noon- string.
1. Innoun_similarity.pyimplement the functionget_similarity_scores(pairs) so that it ranks the pairs in order of decreasing similarity. Hint: the similarity of a pair should be represented by the similarity of the most similar pair of synsets they have. [4 points]
2. In noun_similarity.py implement the function leave_odd_man_out(words) so that it returns the odd word from the given list of words. Hint: use the imple- mented functionget_similarity_scores(pairs). [4 points]
1
Exercise 2: Finding Hyponyms with WordNet [10 points]
In this exercise, you will write a program to find nouns (hyponyms) that belong to certain categories (hypernyms) in wordnet. These categories arerelative,scienceand illness.
Download the file ada_lovelace.txt into the data/ folder of your project. Take a look at the file hw09_wordnet/find_hyponyms.py. Complete some methods to find hyponyms:
1. In the class constructor determine all noun lemmas from ada_lovelace.txt fol- lowing the steps:
• Read text as a string
• Split text into sentences: usenltk.sent_tokenize
• Split sentences into tokens: usenltk.word_tokenize
• Perform POS tagging of tokens
• Lemmatize nouns (any token whose POS tags start with "N"): useWordNetLemmatizer()
• Determine all noun lemmas [6 points]
2. Implement the class method hypernymOf(self,synset1, synset2) by returning True if synset2 is a hypernym of synset1, or if they are the same synsets. Return False otherwise. Hint: use synset1.hypernyms(); do not forget to check whether the hypernym of synset1 is hypernym of synset2 (use recursion). [1 point]
3. Implement the class methodget_hyponyms(self,hypernym). This method should return set of noun lemmas inada_lovelace.txtthat are hyponyms (subordinates) to the hypernym. [3 points]
The output would then look as follows:
Synset: relative.n.01
Lemmas: father, wife, baby, boy, parent, grandchild, son, relation, relative, Family, mother, child, girl, half-sister, daughter, husband
Synset: science.n.01
Lemmas: calculus, phrenology, anatomy, Science, science, government, Magnetism, math, thermodynamics, analysis, mathematics
Synset: illness.n.01
Lemmas: measles, cancer, illness, madness, disease
2
